Method and System for Machine Learning Based Item Matching by Considering User Mindset

1. A processor implemented method for item matching, the method comprising: selecting, by a processor, an item of interest among a plurality of items associated with an entity for identifying a matching item set comprising partial matching items and non-matching items from one of the plurality of items of the entity or a plurality of competitor items of a competitor entity among a plurality of competitor entities having a category identical to the category of the item, wherein each item from the plurality of items of the entity and the plurality of competitor items is defined by a set of Attribute Value (AVs), wherein each AV among the set of AVs correspond to one of a qualitative AV type and a quantitative AV type; performing, by the processor, attribute (AT) enrichment for quantizing a plurality of AVs corresponding to the qualitative AV type based on a Machine Learning (ML) technique, wherein the AT enrichment comprises: (i) determining AV level performance, depicting estimate of sales share per day, for the plurality of AVs of the plurality of items of the entity, corresponding to the qualitative AV type, if the matching item set is to be identified from the plurality of items of the entity; and (ii) determining AV level estimate of price variation per time frame, for the plurality of AVs of the plurality of competitor items corresponding to the qualitative AV type, if the matching item set is to be identified from the plurality of competitor items of the competitor entity, wherein the ML technique comprises training a first ML model to determine the AV level performance using a first training data set extracted from an entity database, wherein the first training data set is a two dimensional data matrix constructed for each category among the plurality of categories, wherein a plurality of columns of the data matrix provide information to the first ML model corresponding to a plurality of sales drivers recorded for the entity and a plurality of rows of the data matrix provide information to the first ML model corresponding to a day for which corresponding sales drivers are recorded, wherein the ML technique comprises training a second ML model to determine the AV level estimate of price variation using a second training data set extracted by crawling data from a plurality of data sources and the entity database, wherein the second training data set is a two dimensional data matrix constructed for each category among the plurality of categories, and wherein a plurality of columns of the data matrix provide information to the second ML model corresponding to the entity, the plurality of competitor entities, and a plurality of ATs of each category and a plurality of rows of the data matrix provide per hour per day information to the second ML model for which data in the plurality of columns is recorded; standardizing, by the processor, values of: the AV level performance of the plurality of AVs of the qualitative AV type of the plurality of items of the entity and the AV level estimate of price variations of the plurality of competitor items, associated with the category of the item; and the plurality of AVs of the quantitative type, corresponding to the plurality of items of the entity and plurality of competitor items, associated with the category of the item; assigning weights, by the processor, to the standardized values, wherein weight is based on Demand Transfer (DT) value provided by a Customer Decision Tree (CDT) obtained for the category of the item, wherein the DT value captures user mindset by capturing percentage shift of demand from one AV to another AV; and identifying, by the processor, the matching item set based on a matching score computed by comparing the weighted standardized values of the plurality of AVs of the item with one of: the weighted standardized values of the plurality of AVs of the plurality of items of the entity, if the matching item set is to be identified from the plurality of items of the entity, wherein the matching item set, arranged in descending order of the matching score, comprises the partial matching items and the non-matching items from the plurality of items of the entity being compared; and the weighted standardized values of the plurality of AVs of the plurality of competitor items, if the matching item set is to be identified from the plurality of competitor items of the competitor entity, wherein the matching item set, arranged in descending order of the matching score, comprises the partial matching items and the non-matching items from the plurality of competitor items of the competitor entity being compared.

2. The method of claim 1 , wherein the first training data set is generated for each segment among a plurality of segments of every segment type among a plurality of segment types defined by the entity.

3. The method of claim 1 , wherein the second training data set is generated for each price zone among a plurality of price zones of every segment type among a plurality of segment types defined by the entity.

4. A system for item matching, the system comprising: a memory storing instructions; one or more Input/Output (I/O) interfaces; and one or more processors coupled to the memory via the one or more I/O interfaces, wherein the one or more processors are configured by the instructions to: select an item of interest among a plurality of items associated with an entity for identifying a matching item set comprising partial matching items and non-matching items from one of the plurality of items of the entity or a plurality of competitor items of a competitor entity among a plurality of competitor entities having a category identical to category of the item, wherein each item from the plurality of items of the entity and the plurality of competitor items is defined by a set of Attribute Value (AVs), wherein each AV among the set of AVs correspond to one of a qualitative AV type and a quantitative AV type; perform attribute (AT) enrichment for quantizing a plurality of AVs corresponding to the qualitative AV type based on a Machine Learning (ML) technique, wherein the AT enrichment comprises: (i) determining AV level performance, depicting estimate of sales share per day, for the plurality of AVs of the plurality of items of the entity, corresponding to the qualitative AV type, if the matching item set is to be identified from the plurality of items of the entity; and (ii) determining AV level estimate of price variation per time frame, for the plurality of AVs of the plurality of competitor items corresponding to the qualitative AV type, if the matching item set is to be identified from the plurality of competitor items of the competitor entity, wherein the ML technique comprises training a first ML model to determine the AV level performance using a first training data set extracted from an entity database, wherein the first training data set is a two dimensional data matrix constructed for each category among the plurality of categories, wherein a plurality of columns of the data matrix provide information to the first ML model corresponding to a plurality of sales drivers recorded for the entity and a plurality of rows of the data matrix provide information to the first ML model corresponding to a day for which corresponding sales drivers are recorded, wherein the ML technique comprises training a second ML model to determine the AV level estimate of price variation using a second training data set extracted by crawling data from a plurality of data sources and the entity database, wherein the second training data set is a two dimensional data matrix constructed for each category among the plurality of categories, and wherein a plurality of columns of the data matrix provide information to the second ML model corresponding to the entity, the plurality of competitor entities, and a plurality of ATs of each category and a plurality of rows of the data matrix provide per hour per day information to the second ML model for which data in the plurality of columns is recorded; standardize values of: the AV level performance of the plurality of AVs of the qualitative AV type of the plurality of items of the entity and the AV level estimate of price variation s of the plurality of competitor items, associated with the category of the item; and the plurality of AVs of the quantitative type, corresponding to the plurality of items of the entity and plurality of competitor items, associated with the category of the item; assign weights to the standardized values, wherein weight is based on Demand Transfer (DT) value provided by a Customer Decision Tree (CDT) obtained for the category of the item, wherein the DT value captures user mindset by capturing percentage shift of demand from one AV to another AV; and identify the matching item set based on a matching score computed by comparing the weighted standardized values of the plurality of AVs of the item with one of: the weighted standardized values of the plurality of AVs of the plurality of items of the entity, if the matching item set is to be identified from the plurality of items of the entity, wherein the matching item set, arranged in descending order of the matching score, comprises the partial matching items and the non-matching items from the plurality of items of the entity being compared; and the weighted standardized values of the plurality of AVs of the plurality of competitor items, if the matching item set is to be identified from the plurality of competitor items of the competitor entity, wherein the matching item set, arranged in descending order of the matching score, comprises the partial matching items and the non-matching items from the plurality of competitor items of the competitor entity being compared.

5. The system of claim 4 , wherein the one or more processors are configured to generate the first training data for each segment among a plurality of segments of every segment type among a plurality of segment types defined by the entity.

6. The system of claim 4 , wherein the one or more processors are configured to generate the second training data set for each price zone among a plurality of price zones of every segment type among a plurality of segment types defined by the entity.

7. One or more non-transitory machine readable information storage mediums comprising one or more instructions which when executed by one or more hardware processors causes a method for: selecting an item of interest among a plurality of items associated with an entity for identifying a matching item set comprising partial matching items and non-matching items from one of the plurality of items of the entity or a plurality of competitor items of a competitor entity among a plurality of competitor entities having a category identical to the category of the item, wherein each item from the plurality of items of the entity and the plurality of competitor items is defined by a set of Attribute Value (AVs), wherein each AV among the set of AVs correspond to one of a qualitative AV type and a quantitative AV type; performing attribute (AT) enrichment for quantizing a plurality of AVs corresponding to the qualitative AV type based on a Machine Learning (ML) technique, wherein the AT enrichment comprises: (i) determining AV level performance, depicting estimate of sales share per day, for the plurality of AVs of the plurality of items of the entity, corresponding to the qualitative AV type, if the matching item set is to be identified from the plurality of items of the entity; and (ii) determining AV level estimate of price variation per time frame, for the plurality of AVs of the plurality of competitor items corresponding to the qualitative AV type, if the matching item set is to be identified from the plurality of competitor items of the competitor entity, wherein the ML technique comprises training a first ML model to determine the AV level performance using a first training data set extracted from an entity database, wherein the first training data set is a two dimensional data matrix constructed for each category among the plurality of categories, wherein a plurality of columns of the data matrix provide information to the first ML model corresponding to a plurality of sales drivers recorded for the entity and a plurality of rows of the data matrix provide information to the first ML model corresponding to a day for which corresponding sales drivers are recorded, wherein the ML technique comprises training a second ML model to determine the AV level estimate of price variation using a second training data set extracted by crawling data from a plurality of data sources and the entity database, wherein the second training data set is a two dimensional data matrix constructed for each category among the plurality of categories, and wherein a plurality of columns of the data matrix provide information to the second ML model corresponding to the entity, the plurality of competitor entities, and a plurality of ATs of each category and a plurality of rows of the data matrix provide per hour per day information to the second ML model for which data in the plurality of columns is recorded; standardizing values of: the AV level performance of the plurality of AVs of the qualitative AV type of the plurality of items of the entity and the AV level estimate of price variations of the plurality of competitor items, associated with the category of the item; and the plurality of AVs of the quantitative type, corresponding to the plurality of items of the entity and plurality of competitor items, associated with the category of the item; assigning weights to the standardized values, wherein weight is based on Demand Transfer (DT) value provided by a Customer Decision Tree (CDT) obtained for the category of the item, wherein the DT value captures user mindset by capturing percentage shift of demand from one AV to another AV; and identifying the matching item set based on a matching score computed by comparing the weighted standardized values of the plurality of AVs of the item with one of: the weighted standardized values of the plurality of AVs of the plurality of items of the entity, if the matching item set is to be identified from the plurality of items of the entity, wherein the matching item set, arranged in descending order of the matching score, comprises the partial matching items and the non-matching items from the plurality of items of the entity being compared; and the weighted standardized values of the plurality of AVs of the plurality of competitor items, if the matching item set is to be identified from the plurality of competitor items of the competitor entity, wherein the matching item set, arranged in descending order of the matching score, comprises the partial matching items and the non-matching items from the plurality of competitor items of the competitor entity being compared.